On Fri, Mar 5, 2021 at 11:41 AM Josh Poimboeuf <jpoimboe@xxxxxxxxxx> wrote: > > On Thu, Mar 04, 2021 at 11:12:42AM -0800, Linus Torvalds wrote: > > On Thu, Mar 4, 2021 at 7:36 AM Masahiro Yamada <masahiroy@xxxxxxxxxx> wrote: > > > > > > All the kernel-space objects are rebuilt > > > when the compiler is upgraded. > > > > I very much NAK'ed that one. Why did that go in? > > > > Or maybe I NAK'ed another version of it (I think the one I NAK'ed was > > from Josh), and didn't realize that there were multiple ones. > > This thread is making me dizzy, Me too. > Ok. So it sounds like the best/easiest option is the original patch in > this thread: when building an external module with a GCC mismatch, just > disable the GCC plugin, with a warning (or an error for randstruct). Just for clarification, I believe "the original patch" pointed to this one: https://lore.kernel.org/lkml/efe6b039a544da8215d5e54aa7c4b6d1986fc2b0.1611607264.git.jpoimboe@xxxxxxxxxx/ This is dead. Please do not come back to this. See negative comments not only from me, but also from Greg, Peter, Christoph. > but I think the patch you NAK'ed from me > was different. It just added an error on GCC mismatch with external > modules: > > https://lkml.kernel.org/r/fff056a7c9e6050c2d60910f70b6d99602f3bec4.1611863075.git.jpoimboe@xxxxxxxxxx > > though I think you were ok with downgrading it to a warning. I think this was not NAKed. At first, Linus NAKed it, but I think it was just because of some misunderstanding. I do not have an objection about checking compiler version difference for external module builds. I was pointing out some mistakes in the Makefile implementation. (I would say "not that way" even with your second trial). -- Best Regards Masahiro Yamada
